Benchmarking de Sistemas AutoML Open-source

Main Article Content

Maria Victória Rodrigues dos Santos
https://orcid.org/0000-0002-3590-2704
Gabriel Mac'Hamilton Renaux Alves
https://orcid.org/0000-0002-3735-190X
Alexandre Magno de Andrade Maciel
https://orcid.org/0000-0003-4348-9291

Abstract

This work offers a benchmarking of the three most know open-source AutoML systems, Auto-WEKA, Auto-Sklearn and TPOT, in terms of how they work in each part of an AutoML flow, and algorithms supported in each part of that flow. Automated Machine Learning is a tool that automates the result of machine learning with as little human effort as possible. This work shows that for certain types of data and forecasting objectives, the users, whether data science students or professionals, must pay attention to each tool, because it implies the results obtained and predictions can be more refined or not.

Downloads

Download data is not yet available.

Article Details

How to Cite
dos Santos, M. V., Alves, G., & Maciel, A. (2022). Benchmarking de Sistemas AutoML Open-source. Journal of Engineering and Applied Research, 7(3), 19-28. https://doi.org/10.25286/repa.v7i3.2456
Section
Edição Especial em Ciência de Dados e Analytics

Most read articles by the same author(s)